search

Ralim / IronOS

Open Source Soldering Iron firmware
https://ralim.github.io/IronOS/
GNU General Public License v3.0
7.28k stars 723 forks source link

TS80 [TS200] Support #349

Closed Ralim closed 1 year ago

Ralim commented 6 years ago

The TS80 runs very similar hardware to the TS100 iron. The firmware does not need many changes to support this.

Changes To do / Progress

Ralim commented 6 years ago

No problem, the firmware has a few rough edges (temp cal isnt perfect, doesnt support >18W mode) but should be usable until i get a full release out soon :)

Houseman74 commented 6 years ago

Thaks pal. Just a few question: what has been changed in the source to enable use of QC3.0 different from the original one? Where is your main page for the new releases of TS-80? I can only see TS-100 one.. why isn't the ST32M drive shown anymore once plugged in? thanks

Ralim commented 6 years ago

Entirely different code :) Basically the TS100 and TS80 are very similar in hardware and as such I extended the TS100 code to work on the TS80 (Thats what this issue here is to track my progress on).

So the code that does the QC3.0 negotiation is totally different to theirs (though inspired by theirs, and using some of the docs on QC they sent me).

The TS80 builds will show up in the releases tab alongside the TS100 firmware when I make a full release when all the features are working

There is no drive anymore, just the menu on the device :)

Basically, this code is build with the free GCC toolchain, which is not as good at optimising as IAR, so way back in the beginning of all this, I took out the virtual drive to make room for extra features instead :) But dont panic, DFU mode still works fine so you can still upgrade etc the same.

gbraad commented 6 years ago

Off-topic; but so far I have only found a QC to PD fast charge converter. I hadn't spend a lot of time on this before, but it seems QC is a very common for the Chinese versions, so very understandable they also choose this for the TS80. I do not agree, but I understand the choice.

Update: Well, I did find this, which is what they call a 'decoy PD/trigger' to convert to several other protocols. Will have to verify, but looks like a device which could work.

joric commented 6 years ago

@gbraad I think one jumper wire (two wires, tops, in case if CC's are not already connected on the rear side of the PCB) and firmware support is much better than a bunch of ugly adapters. Google for example hates QC and deprecates QC support in favor of PD (here).

gbraad commented 6 years ago

to not to take it further off-topic we perhaps would have to discuss this somewhere else. (and yes, I also dislike having to take additional adapters, but soldering inside is also not really preferred).

Ralim commented 6 years ago

Chinese power banks use QC a lot as phones use it, where as PD was really designed for laptops etc.

QC costs about 3 resistors (basically free) to support, where as PD is a much more complicated protocol (1 wire ethernet anyone). PD would require a much larger AFE or a dedicated PD chip. PD isnt really going to be easy to modify into the existing units.

For example, here is ST's descrete AFE :

usb_pd_afe

So if we were to go to doing something like this, it would be better to try and mod a STUSB4500 IC into the unit.

QC is a ugly pain that legally requires you to pay licensing. PD is at least an open standard.

@gbraad Also feel free to start a new issue to discuss options on all of this :)

joric commented 6 years ago

@Ralim There's software implementation by Reclaimer Labs (not sure about circuits needed for this, I thought it's enough to hook up MCU pins to USB pins) but it's 5000 lines of code, including a 1200 line state machine with 29 states that uses 70% of the Arduino program space, hell knows if it's doable but that PD looks like even bigger abomination than QC now :)

Ralim commented 6 years ago

@joric I have the FSUSB interface that code uses. That code impliments the state machine but none of the signalling on the wire, so going with a descrete AFE (like above) requires even more code. Which is why I like the STM IC's instead for PD as they have the state machine inside them (but cost slightly more).

usb PD is a hell of a protocol, and takes a large amount of space if you try and program the state machine etc internally. QC is annoying because of the supposed secrecy around it, but PD is just massively complex because of all the things it supports. (PD is actually a subset of the functions that interface supports).

brianzhouzc commented 6 years ago

Is there any test firmware we can try on? I've been waiting for TS80's support the day it came out :)

gbraad commented 6 years ago

Scroll up and you will find a zip file (a test file was released a few days ago)

On Fri, Sep 28, 2018 at 2:58 PM Brian notifications@github.com wrote:

Is there any test firmware we can try on? I've been waiting for TS80's support the day it came out :)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/Ralim/ts100/issues/349#issuecomment-425341097, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAHZgE5aS34lcAaxsW3brsuP-njQCXPks5ufciigaJpZM4WO3WC .

--

Gerard Braad | http://gbraad.nl [ Doing Open Source Matters ]

brianzhouzc commented 6 years ago

Scroll up and you will find a zip file (a test file was released a few days ago) … On Fri, Sep 28, 2018 at 2:58 PM Brian @.***> wrote: Is there any test firmware we can try on? I've been waiting for TS80's support the day it came out :) — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#349 (comment)>, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAHZgE5aS34lcAaxsW3brsuP-njQCXPks5ufciigaJpZM4WO3WC . -- Gerard Braad | http://gbraad.nl [ Doing Open Source Matters ]

Thanks for the reply. I tried flashing it multiple times and keep getting .ERR file. I'm not using the stock cable, I'm using a XiaoMi USB-C cable which is confirm working on my other devices. Here's the .ERR file if it helps!

TS80.zip

Ralim commented 6 years ago

Hi, That is very common on some computers, this is not the firmware but instead the way Miniware implemented updates. Try coping the file multiple times in a row or using the python loader script etc. This is documented elsewhere on this repo :)

brianzhouzc commented 6 years ago

Hi, That is very common on some computers, this is not the firmware but instead the way Miniware implemented updates. Try coping the file multiple times in a row or using the python loader script etc. This is documented elsewhere on this repo :)

Thanks a lot! I tried again today and its working now. I will report back any issues related to the firmware!

gbraad commented 6 years ago

Off-topic: currently resorting to the following solution:

img_1469

This allows me to use USB-PD as the slim-tip cable requests 20V (this can also be used for the TS100), and the DC-DC converters understands QC and a bunch of other fast charging protocols. This way I can also use any regular ThinkPad slim-tip to power both irons.

gbraad commented 6 years ago

@Ralim is it possible to draw 12V as supply voltage?

Someone reported that using the supplied the supplied silicone rubber cable he experiences noticeable dropouts of the voltage to below 8.4V I can't confirm at the moment

Zyzto commented 6 years ago

@gbraad in the Iron u can set off volt to 13 and I saw video where it reached near 11v so I guess if qc can hit 12v u can

Ralim commented 6 years ago

It's possible to use 12V, the limiting factor is power as most QC devices only support around 18W, but the hardware can do around 25W.

Going to be adding a menu to let you pick which wastage it goes for when selecting a voltage.

It does a tip resistance measurement and uses that to find the correct voltage, as most QC supplies will shut off suddenly if you exceed their power rating.

jvdmeulen commented 6 years ago

Could it be that the tips can't handle more than 18W and therefore it's limited?

Ralim commented 6 years ago

QC3 is typically specified at 18W, which is where that comes from. The tips can work with 24W happily if your charger can handle it. Sadly there is no way in QC world to ask what the charger can do, which is why it defaults to 18W to play it safe. Going to be added in a menu option to bump this to 24W soon.

gbraad commented 6 years ago

18W

Right, even the included powerbrick offers 9V at 2A, or 9-12V at 1.5A. Only slightly more capable powerbricks, like some of the Lenovo Type-C bricks for laptops provide USB-PD and QC with 12V at 2A.

aziraphale commented 6 years ago

Someone reported that using the supplied the supplied silicone rubber cable he experiences noticeable dropouts of the voltage to below 8.4V I can't confirm at the moment

"Someone" = me :) I even use the same profile pic on YT as I do here ;)

My iron reports the voltage dropping to 8.2-8.3V when heating, depending on which supply I use (the QC3 supply that was supplied with the iron made concerning hissing/popping noises when I first plugged it in and it then died on me after only a few minutes' use - something on the secondary side of the transformer has gone short circuit, but I've not pinned down which component is faulty yet; several components measure as dead short, so it seems like a few components in parallel and only one of them has died). Thankfully I was in the middle of recording a video comparing the voltage drops of different supplies when that supply died, so I have actual video evidence of it that I can send to Banggood at some point!

As such, even though 1.5A @ 12V is nominally no more power than 2A @ 9V, we should see less voltage drop along the cable and therefore higher power available at the iron itself.

When powering the iron from my USB power bank, the voltage drops so low when heating that the stock firmware cuts out completely, but I suspect that the voltage at the power bank itself sags in that case. It's not a very powerful QC3 implementation (it can only handle 1.2A at 12V...) so I wouldn't be surprised if 9V drops a bit at 2A. Thankfully your firmware will happily heat at any voltage, otherwise I'd be in the market for a new power bank right now!

I've also confirmed that my iron is quite happy being fed 12V - I used a QC3 trigger module to request 12V from the supply and plugged that straight into the iron. But I think we knew that anyway :)

joric commented 6 years ago

@aziraphale so that was a shitty cable or what? I've noticed data transfer issues on this cable (copying large files from the phone, there were no issues with its stock cable). Sadly it was the only reason to buy the Pro Kit (well, that, and TS-25D tip).

Ralim commented 6 years ago

The cable has thinner wires that I would like, but that helps keep it flexible.

If you want I can add to the loop to keep "hunting" the right voltage so that it can try and calibrate out the wire ?

We can just get 12V 2A easily as the tip is resistive, so current is directly proportional to the voltage.

The unit doesn't measure current sadly, so the control measures the tip resistance and then back calculates the voltage to gain the maximum power.

gbraad commented 6 years ago

used a QC3 trigger module to request 12V from the supply and plugged that straight into the iron.

We knew, but is not the right way. If the software gets modified, it is possible without a trigger module ;-)

Ralim commented 6 years ago

"Works" and "Will trip off the supply" are two different things.If your power supply supports 12V 2A then it will work nicely when I add support for 24W mode. (So be careful if your PSU doesnt support this).

Shouldnt harm anything, but would rather most people dont risk it so I can't be blamed :|

thepowersource commented 6 years ago

Ralim when will the next revision be ready? Also will I be able to get 12V@2A via a 3S pack without the QC? How much power can the hardware cope with safely?

Ralim commented 6 years ago

Currently under a fair but of work so not getting much forward progress. Its currently usable but missing 18/24W selection.

12V it can usually cope with. As far as I know 25W is about the limit safely.

gji commented 6 years ago

12V it can usually cope with. As far as I know 25W is about the limit safely.

This might be a bit offtopic, but what exactly limits the max power? I can't find any schematics anywhere, but I'm guessing it's some IC? I'm thinking about buying a STREF-SCS001V1 (when/if it's ever available...) and powering the TS80 with a 45W supply, but it sounds like I might blow it out...

thepowersource commented 6 years ago

Thanks for the response Ralim. I would be interested in replacing a few components to make this iron higher powered like gji. It would be the ultimate portable iron considering the efficiency of the cartridges. I solder a lot of heavy AWG terminals and cables so the more power the better. I'll be holding off until the next firmware version is ready but am looking forward to see what this iron can really do!

thepowersource commented 6 years ago

It would be good to take the chipset from the TS100 and combine it with the TS80 chassis and cartridges.

Ralim commented 6 years ago

As far as I am aware max power limit is the heater tip + 3.5mm connector used being rated for a maximum power of 2A. The main fet is around 6, the usb-C is 3A. The PCB may not be laid out for more than 2A either but don't know an easy way to confirm this.

Basically the voltage supplied is adjusted to target 18 or 25W. Since there isn't a huge amount that can be done to control the maximum power easily. If you were to use it with a larger power supply I would just fix the output voltage at 12V and you should be fine :)

@thepowersource I think you will rapidly run into the limits of the hardware (tip and it's connector). It's not going to be a simple swap. If you want the higher power you are talking about and the TS80 just isn't enough then I would suggest just going with the ts100.

The TS80 at the 25W mark looks to be quite capable for most "heavy" wire I come across these days (approx 10awg). So long as you can stand to wait a little while 🙃

thepowersource commented 6 years ago

I might just have to accept the TS80 will not be able to work with the larger AWG stuff if it's connector and PCB are limited to 2A and the chip V limited. I'll probably still get a TS80 as it's the most promising iron out at the moment (obviously with your firmware), I'll get a 200W iron for heavy stuff. I hate melting connectors and insulation or putting crap loads of heat into surrounding parts when soldering large AWG cables.

erivek commented 6 years ago

The range of tips for this iron seems kind of limited. Is there any possibility to include support for Wellers RT-type tips?

gji commented 6 years ago

Marco Reps' review indicates that the pinout is different: https://www.youtube.com/watch?time_continue=345&v=83STpUXHilk

Ralim commented 6 years ago

The pinout for Weller RT tips are indeed different and are completely non-compatible. Might be possible to hardware mod these to work but i'm unsure at the moment if that would work.

Ralim commented 6 years ago

Merged the initial support into master.

Throwing a build here to hopefully nail down any issues. Please let me know if you have issues with the QC3.0 as I only have 2 devices to test with. Also would love any QC2 feedback. TS80.zip

Running with a tight ~700 bytes free in the flash.

gbraad commented 6 years ago

Will test... still waiting for some QC2.0 devices to arrive to test against. Unfortunately the seller on taobao got ill and delayed to ship.

Houseman74 commented 6 years ago

Hi Ralim. I will also test it tonight. Will let you know. Thanks for the great work!!:blush:

whitehoose commented 6 years ago

Got my iron yesterday - worked through your september release last night quickly went through the menus (I really struggle using them cleanly - I think it's just a case of lack of practice) and everything seems good. Couple of option to calibrate not functional, but you know that (couldn't be bothered boiling water etc). But I like the general layout.

Just loaded October's offering, and it somehow feels less versatile. Cycling through yes/no is fine, but when you see the desired temp whizz by for the 5th time it's not so OK. Having said that its good to be able to use (what looks like) just about any power source in an emergency is great. My worst 1A powerbank melted solder which could save me a long walk someday.

Overall I like your style to date and I look forward to the next offering. Thanks for your efforts

jvdmeulen commented 6 years ago

image It's a bit mucht to the right. The display is showing it, if i peek from from an angle. I don't know if my display is misaligned.

PaulGD03 commented 6 years ago

@jvdmeulen iirc the display on the TS80 is shifted to the right from factory. I think you can open it and move it manually

jvdmeulen commented 6 years ago

@PaulGameDev ah, that's the quick fix then. But if everyone has that 'problem' maybe it's better to correct it in the code. I don't think everyone is going to open the iron and adjust it (even tho it's little to no effort)

MaienM commented 6 years ago

The text on my TS80 is aligned properly, so this doesn't seem to be an universal issue.

PaulGD03 commented 6 years ago

might be an issue that only occurs on certain product charges

whitehoose commented 6 years ago

I've got that same screen in the display now and mine is slap in the middle. I might just be lucky. Are people saying the screen has physically slipped - or that the display is showing a mis-aligned image? I'm not clear. Don't know what a run comprises in numbers but the vids show the screen is mounted over components and doesn't look to have a mount or fixing so I''d guess its mechanical slippeage and not software. and that could be isolated to people who have opened the case - or down to units assembled by 1 worker. Shame for the average punter who is reluctant to disassemble - but for those contemplating mods - it should be seen as a bonus.

Ralim commented 6 years ago

Calibration should now be functional i think in the latest build. By default it will show you just the simple offset cal function, but if you change the tip type to custom then you can do an advanced cal using a tip thermocouple or simple using boiling water.

The screen issue is mechanical and doesnt seem to be overly common, Not much I can do about it really 🤷‍♂️ Its mostly held in place by the case pressing on it against the foam behind it.

Houseman74 commented 6 years ago

Ralim: New firmware issues (at least in my device):

Once in setting menu how to exit to soldering?

Now ONE GREAT feature is that drawing at 5V 0.8A the TS-80 can reach the temperature (even if in slower time) so from a normal 5V powerbank the unit can be used!!!! But as usual powermeter shows Unknown mode, 5.12V (D+ 0.94V D- 2.77V) and TS-80 shows 7.5V Hope this helps

Ralim commented 6 years ago

Hi, Can you get a picture of the off screen issue ?

Interesting that your TS80 reads the input voltage rather differently to mine, Ill investigate this more here before saying too much :)

The power bank going into sleep is a known issue, its just what power banks do. Thinking about making so that it "pulses" the tip on for like 100ms every so often just to try and keep the bank on. Not sure about this one. Thoughts ? The risk is a slowly warming tip when not in use.

The QC2/QC3 detection I have found to be rather hit and miss on most of these units :( Since QC2 and QC3 mostly differ in exact voltage levels, but then most power banks dont even follow these correctly. Also im implimenting my timing slightly off spec which in all of my testing made it much more tolerant to lower quality QC devices.

Could you try doing a factory reset (advanced menu) and checking the voltages on screen again ? 😄 Your voltages are off by almost perfect multipler so just wondering if a setting is out of whack :)

Yes I agree! on 5V its slow but good in a pinch if something goes wrong.

Houseman74 commented 6 years ago

Ralim you are right. After factory reset the voltage shows correctly in idle mode (now during heating up it is slightly off by -0.5V. PMETER shows 5V / TS-80 shows 4.5V - must be related to the wire resistance). Factory reset fixed also the screen issue. GREAT!! I would recommend it to everyone has display issues. You are also right in QC2.0/QC3.0 , I have seen it is more than try and hit. Different times plugging in into QC3.0 PORT not always enables the feature- once it stays in 5V mode, another time it jumps to 9V (but never reached 12V!!) One thing I noted is: QC2.0 mode: the inglorious "Unknown mode" is displayed on the power-meter once D- value goes above 0.81V (during heating up) or below 0.59V (during sleeping mode) (D+ stays fixed to 3.0V) below that value (oscillating from 0.70V to 0.8V) it stays fixed QC2.0 mode.

Ok for power-bank to go to sleep but now it goes off also during warming up. In previous FW this does not happen. Regards and thanks for the great effort.